274
BIOCHEMICAL SOCIETY TRANSACTIONS
The Relation of Bile Proteins to Serum and Liver Plasma Membrane
BARBARA M. MULLOCK and RICHARD H. HINTON
Wolfson Bioanalytical Centre, University of Surrey, Guildford GU2 5 XH, Surrey, U.K.
Although there have been many studies on the low-molecular-weight constituents of
mammalian bile, there is no clear agreement on the origin of bile proteins. An obvious
source is the outer leaflet of the hepatocyte plasma membrane, and indeed it has been
shown that bile is rich in enzymes typical of the plasma membrane (Holdsworth &
Coleman, 1975; Evans et a/.,1976) and that bile salts are capable of solubilizing membrane proteins (Vyvoda et al., 1977). Moreover, it is clear that the 5’-nucleotidase of bile
is immunologically identical with the enzyme present in liver plasma membrane (Evans
et a/., 1976; Mullock P t al., 1977). However, immunological studies have also shown the
presence of serum proteins, notably albumin and immunoglobulins, in human bile
(Russel & Burnett, 1963). A protein with the immunological properties of serum
albumin is also present in rat bile, but its electrophoretic mobility in sodium dodecyl
sulphate-containing polyacrylamide gels differs from that of native serum albumin
(Evans et a/., 1976). Further, the general pattern of rat bile proteins is quite distinct from
that of serum proteins (Hinton & Mullock, 1977).
Thus data available at present do not support a single source for bile proteins. It
would seem clear that there is some solubilization of proteins, including enzymes such
as 5’-nucleotidase, from the bile-canalicular face of hepatocytes, but that such solubilization cannot explain the presence of serum proteins in bile. Two hypotheses that
would explain this observation are: (a) the ‘mistaken’ discharge of secretion granules
containing serum proteins at the bile-canalicular rather than at the sinusoidal face of
hepatocytes; (6) the discharge of serum proteins taken up and processed through the
lysosomal system, for it has been suggested by deDuve & Wattiaux (1966) that lysosomes
discharge their contents into the bile canaliculi. We have now attempted a systematic
study of the immunological and electrophoretic properties of the proteins of rat bile
with a view to improvingour understanding of their origin.
Hooded rats of the University of Surrey strain were used in all experiments. Blood and
bile were collected as described earlier (Mullock et al., 1977). Crossed immunoelectrophoresis using single or split antibody-containing gels was carried out as described by
Axelsen e t a / . (1973). Anti-(rat serum) was from Mercia (Watford, Herts., U.K.). Anti(cholestatic rat serum) and anti-(liver plasma membrane) antisera were prepared by
using the dosage schedule given by Mullock et al. (1977). Two dimensional electrophoresis with the first dimension run in agarose gel and the second dimension in a
gradient polyacrylamide gel, was performed as described by Hinton & Mullock (1977).
Crossed immunoelectrophoresis of rat bile proteins against anti-(rat serum) and anti(liver plasma membrane) revealed the presence of at least 14 proteins (Table 1). No
difference was found between the reactions of bile with anti-(rat serum) and anti(cholestatic rat serum). Of the 14 proteins only one, a relatively minor component,
reacted with anti-(liver plasma membrane) and not with anti-(rat serum). Three proteins
reacted with anti-(rat serum) and with all preparations of anti-(liver plasma membrane) ;
another protein reacted with anti-(rat serum) and with some anti-(liver plasma membrane) preparations. The relative strength of the reaction of all four proteins with the
different antisera suggested that all were, in fact, serum proteins [our anti-(liver plasma
membrane) antisera contain antibodies to a number of serum proteins (Issa et al.,
1977)l. The remaining nine bile proteins reacted with anti-(rat serum), but not with
anti-(liver plasma membrane). Bile lipoprotein (Manzato et al., 1976) did not appear to
react with anti-(rat serum) or with anti-(liver plasma membrane), for none of the peaks
obtained by crossed imrnunoelectrophoresis stained with Sudan Black B.
Only nine proteins could be distinguished after two-dimensional electrophoresis of
rat bile proteins. By measurement of the electrophoretic mobility in the first (agarose)
dimension and by absorption of the bile with immobilized anti-(rat serum) it was
possible to correlate most components identified on two-dimensional electrophoresis
1978
1
Mobility*
+
+
+
+
+
+
+
+
+
+
+
+
+
Reaction with
anti-(rat serum)
+
+-
Reaction with
anti-(liver plasma
membrane)
Identified by crossed immunoelectrophoresis
No.
A
1
Mobilityt
+
Staining with
periodic acid/
Schiffs reagent$
+
+
+
+
+
+
+
+
2 x 105
>lo6
>lo6
>lo6
2 x 105
105
2 x 105
5 x 104
Approx. mol.wt.§
7 x 104
Identified by two-dimensional electrophoresis
0.88
?
B
0.82
0.82
4
0.78
C
0.74
5
0.76
6
0.66
7
0.66
DII
0.61
8
0.56
9
0.54
E
0.49
10
0.49
F
0.29
11
0.36
G
0.23
12
0.29
H
0.17
13
0.23
I
0.10
?
14
0.17
* Relative to protein 1 (serumalbumin).
t Movement'in the first (agarose) dimension relative to protein A (equivalent to protein 1).
$ All proteins which failed to stain with periodic acid/Schiffsreagent stained only weakly with Coomassie Brilliant Blue.
0 Estimated by comparison with the position of serum proteins run on similar plates.
!I Protein D does not react with either antiserum.
No.
1
2
3
r
Table 1 . Properties of the major proteins of rat bile
B
I!
r
0
8
m
m
3
276
BIOCHEMICAL SOCIETY TRANSACTIONS
with proteins identified o n crossed immunoelectrophoresis (Table 1 ) . All the major
components separated by two-dimensional electrophoresis were stained by the periodic
acid/Schiff reaction (Smith, 1976), but the staining of component I, which is identified
as the protein containing anti-albumin antigens, stained relatively weakly as
compared with its staining with Coomassie Brilliant Blue R.
We mentioned above three possible sources for bile proteins, namely solubilization
of the bile-canelicular plasma membrane, ‘mistaken’ discharge of secretion granules
at the bile-canelicular rather than at the sinusoidal surface of the hepatocyte, and,
finally, discharge of the contents of lysosomes. Our results show clearly that
only a small proportion of the proteins of bile can derive from hepatocyte plasma
membranes, whereas a large proportion of bile proteins are immunologically
identical with serum proteins. However, the pattern of bile proteins o n two-dimensional
electrophoresis and on crossed immunoelectrophoresis differs t o o much from that
obtained with rat serum for our results t o be explained by contamination of bile by
serum. It would also seem unlikely that the serum proteins in bile could derive from
secretion granules, as bile contains immunoglobulins, which are made outside liver, but
not macroglobulins, which are made in hepatocytes. Hence it would seem possible that
the serum proteins that we recognize in bile are those which have survived passage
through lysosomes, and indeed limited degradation would explain the observation of
Evans et al. (1976) that the bile protein with the immunological properties of albumin
migrates anomalously on electrophoresis in sodium dodecyl sulphate-containing
polyacrylamide gels. It should, however, be noted that bile contains proteins that react
neither with anti-(rat serum) or anti-(liver plasma membrane), but it is not clear whether
this is due t o true immunological difference or t o making of antigenic sites by bound
lipid or bile salts.
Financial support was provided by the Medical Research Council.
Axelsen, N. H . , Kroll, J. & Weeke, B. (1973)A Manual of Quantitative Ininiunoelectrophoresis
Universitetsforlaget, Oslo
de Duve,C. & Wattiaux,R.(1966) Annu. Rev. Physiol. 28,435-492
Evans, W. H.,Kremmer,T. &Culvenor,J. G.(1976)Biochem.J.154,589-595
Hinton,R. H . &Mullock,B. M.(1977)C/in.Chim. Acfa,78,159-162
Holdsworth, G. &Coleman, R. (1975)Biochim. Biophys. Acta389,47-50
Issa, F. S.,Mullock, B. M., Dobrota, M. & Hinton, R. H. (1977) Methodol. Deu. Biochem. 6 ,
171-184
Manzato, E., Fellin, R., Baggio, G., Walch, S.,Neubeck, W. & Seidel, D. (1976)J . Clin. Inwsr.
57,12484 260
Mullock,B. M.,Issa,F. S . &Hinton,R. H.(1977)Clin. Chim. Acta79,129-140
Russel,I. S.&Burnett, W. (1963) Gasfroenterology45,730-739
Smith, I. (1976) in Chromatographic and Electrophoretic Techniques (Smith, I., ed.), 2nd edn.,
pp. 21 1-249, Heinmann, London
Vyvoda, 0.S.,Coleman, R. &Holdsworth, G. (1977)Biochim.Biophys Acta. 465,68-76
Preparation and Properties of the Polypeptides of the Gap Junction
J. CARREIRA and W. H. EVANS
National Institute for Medical Research, Mill Hill, London NW7 I A A , U.K.
In most tissues, the plasma membrane contains discrete areas specialized for the exchange
of small molecules between adjacent cells. These specialized membrane regions, designated gap or communicating junctions (maculae communicantes), have a highly characteristic morphology fully in keeping with their role in forming transmembrane channels
between the interiors of coupled cells (Casper et al., 1977; Makowski et al., 1977).
Studies have shown that a variety of physiological molecules of mol.wt. u p to 1800 can
1978